it feels to me like this is two levels of abstraction above what's necessary to implement those algorithms, which is again like two or three levels of abstraction above what's necessary to actually use them...
The best line of all time. Imagine being a TA for the calculus II class in which the prof. put that on the exam. I can hear the trenchant complaints already.
![30 ชั่วโมงคนเดียว บนรถไฟชั้น 3 โซเวียต [Around The World Ep. 2]](http://i.ytimg.com/vi/MlE9boAVqDA/mqdefault.jpg)
Very mathematical, but at the same time cleared the whole set of stochastic processes and the nonparametric toolbox for further use! Thanks!
The first 10 minutes REALLY need to be the standard storytelling
it feels to me like this is two levels of abstraction above what's necessary to implement those algorithms, which is again like two or three levels of abstraction above what's necessary to actually use them...
Such a great talk!
Is there some kind of general approach to finding a measure like what's done at 5:05 for other similar problems other than guessing?
Bang, bang, and bang. Then throw that in there, and you got an integral that an undergrad can do.
...
*nervous silence from the audience*
The best line of all time. Imagine being a TA for the calculus II class in which the prof. put that on the exam. I can hear the trenchant complaints already.
Wtf is a beta 1 com alpha distribution? Or what does he say at 8:00
Beta(1,alpha)
1/B(1, alpha) (1 - eta)^(alpha-1) is the probability density function of Beta(1, alpha).